Light activation of protein splicing with a photocaged fast intein

Article abstract of DOI:10.1021/ja508597d

Intein-mediated protein splicing has found broad biotechnological applications. Herein, we describe our recent result in engineering a photoactivatable intein compatible with living mammalian cells. A photocaged cysteine amino acid residue was genetically introduced into a highly efficient Nostoc punctiforme (Npu) DnaE intein. The resulting photocaged intein was inserted into a red fluorescent protein (RFP) mCherry and a human Src tyrosine kinase to create inactive chimeric proteins. A light-induced photochemical reaction was able to reactivate the intein and trigger protein splicing. Active mCherry and Src were formed as observed by direct fluorescence imaging or imaging of an Src kinase sensor in mammalian cells. The genetically encoded photocaged intein is a general optogenetic tool, allowing effective photocontrol of primary structures and functions of proteins.

Full text of DOI:10.1021/ja508597d

Communication
pubs.acs.org/JACS
Light Activation of Protein Splicing with a Photocaged Fast Intein
†
†
Wei Ren, Ao Ji, and Hui-wang Ai*
Department of Chemistry, University of California, Riverside, 501 Big Springs Road, Riverside, California 92521, United States
*
S Supporting Information
The Nostoc punctiforme (Npu) DnaE intein is among the
most well-characterized and efficient inteins, with a splicing
ABSTRACT: Intein-mediated protein splicing has found
broad biotechnological applications. Herein, we describe
our recent result in engineering a photoactivatable intein
compatible with living mammalian cells. A photocaged
cysteine amino acid residue was genetically introduced into
a highly efficient Nostoc punctiforme (Npu) DnaE intein.
The resulting photocaged intein was inserted into a red
fluorescent protein (RFP) mCherry and a human Src
tyrosine kinase to create inactive chimeric proteins. A light-
induced photochemical reaction was able to reactivate the
intein and trigger protein splicing. Active mCherry and Src
were formed as observed by direct fluorescence imaging or
imaging of an Src kinase sensor in mammalian cells. The
genetically encoded photocaged intein is a general
optogenetic tool, allowing effective photocontrol of
primary structures and functions of proteins.
1
2,13
reaction half-life of ∼60 s at 37 °C.
The Npu DnaE intein is
also compatible with a myriad of flanking extein sequences.
14
All these features make the Npu DnaE intein an ideal research
tool, especially for mammalian studies. Mutagenesis of the first
catalytic cysteine residue within the Npu DnaE intein to alanine
(
Cys1Ala) abrogates protein splicing and autocleavage at both
12,15
intein domain ends.
This property is different from that of
some other recently reported fast inteins, whose Cys1Ala
mutants are efficient in undergoing the C-terminal cleavage
16
reaction.
The genetic code expansion technology is capable of
introducing site-specific photocaged lysine, tyrosine, serine,
and cysteine residues into proteins of interest in living systems,
1
7−21
including bacterial, yeast, and mammalian cells.
Previously,
22−24
25
optical control of enzymatic activities,
ion channels, gene
expression and silencing, and protein translocation
26
27,28
have
been demonstrated by replacing critical protein residues with
photocaged unnatural amino acids (UAAs). In this study, we
show that a genetically encoded photoactivatable intein can be
readily derived by replacing the Cys1 residue of Npu DnaE
intein with a photocaged cysteine, and it is highly effective in
directly modulating primary protein structures, thereby
rendering a general approach for controlling protein activities
in living cells.
nteins are protein elements that are capable of excising
I
themselves and subsequently splicing adjacent N- and C-
1
terminal extein flanks to form a new truncated peptide. These
naturally occurring, self-catalyzing protein-splicing elements
have been adapted to achieve efficient protein purification,
2
,3
ligation, labeling, cyclization, cleavage, and patterning. In
particular, conditional inteins, whose activities are inducible by
additional factors, such as small molecules, light, or changes in
temperature, pH, or redox states, have previously been utilized
Previous efforts have utilized mutant pairs of pyrrolysyl-
2
9,30
tRNA synthetase (PylRS)/tRNA
and Escherichia coli leucyl-
4,5
25
to regulate protein activities in vitro and in vivo. Photo-
activatable inteins are of particular interest because light-based
approaches often have sufficient spatial and temporal resolution
to meet the need of understanding biology at the cellular and
tRNA synthetase (EcLeuRS)/tRNA in mammalian cells for
the genetic encoding of unnatural cysteine derivatives that can
be decaged with long-wavelength UVA radiation. In partic-
ularly, an orthogonal EcLeuRS/tRNA pair originally engineered
19
6
for the encoding of a photocaged serine in yeast was found to
subcellular levels. In a previous work, Noren et al. reported the
be capable of encoding a photocaged cysteine (1 in Figure 1a)
in vitro preparation of a photoactivatable Thermococcus litoralis
25
in mammalian cells. Based on these results, we modified our
(
Tli) Pol-2 intein, using a chemically aminoacylated suppressor
31
7
pMAH mammalian expression plasmid to express the mutant
EcLeuRS and tRNA genes. Expression of the full-length GFP
protein in Human Embryonic Kidney (HEK) 293T cells
bearing EGFP-Tyr39TAG (a gene for enhanced green
fluorescent protein with an amber codon at residue 39) was
observed to be dependent on 1 (Figure 1b). Photolysis of 1 is
expected to generate an aldehyde byproduct, which may further
react with free cellular amines to inadvertently promote cell
tRNA. Furthermore, chemical synthetic methods have also
been employed to integrate photocleavable functional groups
8
9
into the O-acyl isomer, the peptide backbone, or the N-
1
0
terminus of split inteins to achieve photocontrolled protein
splicing. Due to the difficulty of directly delivering proteins or
peptides into living cells, these studies focused on in vitro
applications. In another work, two photoresponsive dimeriza-
tion domains were each fused to an artificially split intein
fragment as a genetically encoded system to control protein
splicing in living Saccharomyces cerevisiae cells, but the system
32
toxicity (Figure S1a). Therefore, we also prepared a new
UAA, 2 (Figure 1a), photolysis of which yields a cysteine and a
less reactive ketone byproduct (Figure S1bc). Since 2 is
structurally similar to 1, we also tested 2 for amber suppression
11
was not adaptable to mammalian cells. Herein, we report the
genetic encoding of a photoactivatable intein and its
applications in directly controlling primary structures of
proteins and therefore their functions, in living mammalian
cells.
Received: August 21, 2014
©
XXXX American Chemical Society
A
DOI: 10.1021/ja508597d
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
was comparable to the rate of chromophore maturation of
34
mCherry. These results indicate that the caged intein was
photoactivated to undergo protein splicing and form a highly
fluorescent reconstituted mCherry. Since the construct was
6
xHis-tagged at the C-terminal end, Ni-NTA agarose beads
were utilized to purify proteins from untreated or UVA-treated
cells. SDS-PAGE analysis of the proteins confirmed the highly
efficient, light-induced protein splicing: upon UVA-treatment,
nearly all of the chimeric protein was converted to the spliced
product (Figure 2c).
We next explored the use the photocaged intein in
controlling enzymatic activities. We inserted the photocaged
intein into the catalytic domain of Src, a human tyrosine
kinase. The kinase catalytic domain has eight cysteine
residues and 12 serine residues. We designed chimeric proteins
by randomly and individually inserting the intein into three
sites in Src (Figure 3a and Figure S4). First, we inserted the
Figure 1. Genetic encoding of photocaged cysteines in HEK 293T
cells. (a) Chemical structures of two photocaged cysteines, 1 and 2.
35
(
b) SDS-PAGE analysis of Ni-NTA-purified EGFP, containing 1 or 2,
expressed in HEK 293T cells. (c) Microscopic imaging of EGFP-
expressing HEK 293T cells in the absence (left column) or presence
(
right column) of 2 (scale bar: 50 μm).
in the presence of the mutant EcLeuRS/tRNA pair. We
achieved an appreciable yield of full-length GFP from HEK
293T cells, as observed by SDS-PAGE analysis and fluorescence
microscopic imaging (Figure 1b and c). Electrospray ionization
mass spectrometry (ESI-MS) further confirmed the genetic
incorporation of 2 in the recombinantly expressed EGFP
(
Figure S2).
To determine whether 2 can be utilized to photocontrol the
protein splicing activity of the Npu DnaE intein, we inserted a
full-length Npu DnaE intein sequence into mCherry (Figure
2a). The residue 138 on a long loop between the β-strands 6
Figure 3. Photoactivation of Src kinase. (a) Primary structures of the
chimeric proteins tested in this study. The gray portion of the bars
represents the sequence of the human Src kinase between the
indicated residues. The asterisk (*) indicates the Cys1 residue for
UAA incorporation; “M” is methionine, as the translational start site;
and “C” is cysteine, used to replace residue 342 of Src. (b) Activity of
the chimeric proteins before and after UVA irradiation, as measured
from FRET ratios of a KRas-Src sensor. In the absence of 2, the full-
length proteins were not synthesized and are thus used as negative
controls. A wild-type Src was also prepared as a positive control. To
block ribosomal protein synthesis during and after UVA irradiation,
cycloheximide (CHX) was also added to a control group. (c)
Pseudocolored ratio images of representative UVA-treated HEK 293T
cells expressing the F1 construct in the presence of 2 at the indicated
post-treatment time (in minutes). The color bar represents
fluorescence ratio (YPet/ECFP) (scale bar: 25 μm). (d) FRET ratios
plotted versus time for HEK 293T cells. Color symbols are for
individual cells in panel c, marked at 0 min by arrows in the same
colors. The FRET ratios of an identically treated control cell cultured
in the absence of 2 (see Figure S5) are shown as open black circles.
Figure 2. Photoactivation of mCherry. (a) Primary structures of the
intein/mCherry chimeric protein and its photoconverted product after
UV-induced protein splicing. The red portion of the bar represents the
mCherry sequence. The asterisk (*) represents the Cys1 residue for
UAA incorporation. The “CM” region are two extein residues (+1 and
+
2). (b) Microscopic imaging of HEK 293T cells expressing the
construct treated with or without UV irradiation (scale bar: 50 μm).
c) SDS-PAGE analysis of the Ni-NTA-purified proteins from HEK
93T cells, with or without UV irradiation.
(
2
and 7 of mCherry was chosen as the insertion site (Figure
3
3
S3). Moreover, the codon of the Cys1 residue of Npu DnaE
intein was mutated to an amber codon (TAG) for UAA
incorporation. The chimeric construct was subsequently
expressed in HEK 293T cells, with cell culture media
containing 2. Almost no fluorescence was observed prior to
UVA treatment (Figure 2b), suggesting that the intein insertion
disrupted the fluorescence of mCherry. Next, we used a UVA
lamp to directly illuminate cells in cell culture dishes, and
strong red fluorescence was observed in 1 h after irradiation
intein between Gly276 and Cys277, or Val399 and Cys400 of
Src (F1 and F2 in Figure 3a). For these two constructs, protein
splicing is expected to generate a product identical to the wild-
type Src kinase catalytic domain. We also built the third
construct, F3, in which the intein was placed downstream of
Met341 (Figure 3a). Because the Npu DnaE intein requires a
cysteine residue at the +1 site for efficient protein splicing, we
also mutated Ser342 to cysteine, to which appended was the
native Src sequence from residue 343 to residue 533. The
splicing product of F3 is expected to be different from the wild-
12
(
Figure 2b). This rate of developing red fluorescence in cells
B
DOI: 10.1021/ja508597d
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
type protein by a single Ser342Cys mutation. It is worth noting
that a serine-to-cysteine mutant is tolerated in many cases
the Cys1 residue of a highly efficient Npu DnaE intein. By
incorporating the photocaging group, the protein splicing
activity of the intein was effectively and efficiently inhibited, and
the activity was only observed after a brief exposure to long-
wavelength UVA light. The resulting photocaged intein was
inserted into other proteins to directly control their primary
structures. Because the Npu DnaE intein is compatible with a
myriad of extein sequences, such manipulation should be quite
versatile. A downstream C-extein Cys+1 residue is required for
protein splicing, but cysteine can be found in many proteins. In
addition, a single cysteine mutation may be tolerated by many
proteins. Thus, the approach described here may be applied to
a large percentage of proteins. We acknowledge that additional
N- and C-terminal extein sequences might affect the kinetics of
protein splicing. This issue can be addressed by using evolved
inteins that splice with higher efficiency at various splice
junctions. One might also prepare several chimeric constructs
at different splice sites to screen for variants retaining excellent
expression, stability, and postphotoactivation splicing kinetics.
The use of the photoactivatable inteins to control protein
activity is highly attractive, because it requires little information
on the biochemistry or 3D structures of the proteins of interest.
The photoactivatable intein reported here is a new and
powerful addition to the mammalian optochemical genetic
toolbox, permitting the modulation of proteins directly at the
amino acid sequence level.
36
without dramatically affecting protein activities. We also fused
mCherry at the C-terminal end as an expression indicator of the
UAA-containing full-length proteins. Next, we used a KRas-Src
3
7
̈
sensor, based on Forster resonance energy transfer (FRET)
between ECFP and YPet, to evaluate the activities of F1, F2,
and F3 in the presence or absence of UVA irradiation. This
sensor was well-validated in previous studies, and Src kinase
activity is known to decrease the intensity ratio (YPet/ECFP)
of the sensitized YPet fluorescence emission to the direct ECFP
3
7
donor emission. HEK 293T cells containing each of the 3
constructs and the Ras-Src sensor were treated with UVA light
and, then, lysed for fluorescence quantification with a plate
reader (Figure 3b). All of our three constructs were inactive
prior to UVA irradiation, while UVA light was able to activate
them, leading to the decrease of the FRET ratios of the sensor.
A reduced FRET ratio was also observed for cells coexpressing
a wild-type Src kinase and the Src sensor. Furthermore,
negative control experiments were performed with HEK 293T
cells containing each of the three constructs but cultured in the
absence of 2. Cells in the negative groups were also subjected
to the identical UVA treatment, so that the partial photo-
bleaching of the Src sensor did not mask the FRET changes
caused by the photoactivation of the Src kinase activity.
Moreover, we utilized fluorescence microscopy to closely
monitor the process (Figure 3c). HEK 293T cells coexpressing
the Src sensor and the chimeric F1 construct were irradiated on
an epi-fluorescence microscope equipped with a DAPI
excitation filter. Next, we carried out time-lapse, two-channel
FRET imaging of ECFP and YPet. The FRET ratios of the Src
sensor gradually decreased in the monitored 30 min period. In
contrast, the UVA-treated control cells cultured in the absence
of 2 showed no obvious change in FRET ratios during the
imaging period (Figure 3d and Figure S5). It was noted that
considerable Src-induced FRET changes occurred during the 2
min of UVA illumination. Analysis of single cells showed that
the average FRET ratio (YPet/ECFP) at 0 min, when time-
lapse FRET imaging started, was 2.11 ± 0.08 for cells
containing the photoactivated Src. In comparison, negative
cells identically treated with UVA radiation had an average
FRET ratio of 2.35 ± 0.03. This is not surprising, considering
the fast kinetics of the Npu DnaE intein. The UVA illumination
39
ASSOCIATED CONTENT
Supporting Information
■
*
S
General methods, protein sequencing information, mass
spectrometry, and FRET imaging of control cells. This material
is available free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
huiwang.ai@ucr.edu
■
*
Author Contributions
†
W.R. and A.J. contributed equally to this work.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
38
condition did not affect cell viability but effectively activated
the photocaged intein to promote the formation of Src via
protein splicing. These data support that the photocaged Npu
DnaE intein is an effective tool for the control of enzyme
activities.
This work was support by the University of California,
Riverside. We acknowledge Philip Lee for technical assistance,
Tan Truong for helping with manuscript preparation, the UCR
Analytical Chemistry Instrumentation Facility for sample
analysis, and Prof. Hideo Iwai (University of Helsinki), Prof.
William Hahn (Harvard), and Prof. Yingxiao Wang (UCSD)
for providing several plasmids used in the work.
UV radiation may also decage the charged unnatural
aminoacyl tRNA, which may be further utilized by cellular
ribosomes to synthesize proteins. We added cycloheximide
REFERENCES
(100 μg/mL) to block ribosomal protein synthesis during and
■
after irradiation, the photoactivation of Src kinase was not
affected (Figure 3b). In addition, the activation of Src was
observed right after UV irradiation (Figure 3d), when
ribosomal protein synthesis from the decaged aminoacyl
tRNA was unlikely to be achieved in this short time frame.
These results suggest that the direct decaging of the
accumulated chimeric proteins in cells was the major pathway
in our experiments.
(1) Hirata, R.; Ohsumk, Y.; Nakano, A.; Kawasaki, H.; Suzuki, K.;
Anraku, Y. J. Biol. Chem. 1990, 265, 6726.
(
(
(
(
(
2) Shah, N. H.; Muir, T. W. Chem. Sci. 2014, 5, 446.
3) Topilina, N. I.; Mills, K. V. Mobile DNA 2014, 5, 5.
4) Mootz, H. D. ChemBioChem. 2009, 10, 2579.
5) Peck, S. H.; Chen, I.; Liu, D. R. Chem. Biol. 2011, 18, 619.
6) Toettcher, J. E.; Voigt, C. A.; Weiner, O. D.; Lim, W. A. Nat.
Methods 2011, 8, 35.
7) Cook, S. N.; Jack, W. E.; Xiong, X.; Danley, L. E.; Ellman, J. A.;
(
In summary, we have engineered the first genetically encoded
photoactivatable intein compatible with living mammalian cells,
in which a photocaged cysteine is used to genetically replace
Schultz, P. G.; Noren, C. J. Angew. Chem., Int. Ed. 1995, 34, 1629.
(8) Vila-Perello, M.; Hori, Y.; Ribo, M.; Muir, T. W. Angew. Chem.,
Int. Ed. 2008, 47, 7764.
C
DOI: 10.1021/ja508597d
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
(
1
(
2
9) Berrade, L.; Kwon, Y.; Camarero, J. A. ChemBioChem. 2010, 11,
368.
10) Binschik, J.; Zettler, J.; Mootz, H. D. Angew. Chem., Int. Ed.
011, 50, 3249.
(
11) Tyszkiewicz, A. B.; Muir, T. W. Nat. Methods 2008, 5, 303.
12) Zettler, J.; Schutz, V.; Mootz, H. D. FEBS Lett. 2009, 583, 909.
13) Ellila, S.; Jurvansuu, J. M.; Iwai, H. FEBS Lett. 2011, 585, 3471.
14) Cheriyan, M.; Pedamallu, C. S.; Tori, K.; Perler, F. J. Biol. Chem.
(
(
(
2
013, 288, 6202.
15) Ramirez, M.; Valdes, N.; Guan, D.; Chen, Z. Protein Eng. Des.
Sel. 2013, 26, 215.
16) Carvajal-Vallejos, P.; Pallisse, R.; Mootz, H. D.; Schmidt, S. R. J.
Biol. Chem. 2012, 287, 28686.
17) Wu, N.; Deiters, A.; Cropp, T. A.; King, D.; Schultz, P. G. J. Am.
Chem. Soc. 2004, 126, 14306.
18) Chen, P. R.; Groff, D.; Guo, J.; Ou, W.; Cellitti, S.; Geierstanger,
B. H.; Schultz, P. G. Angew. Chem., Int. Ed. 2009, 48, 4052.
19) Lemke, E. A.; Summerer, D.; Geierstanger, B. H.; Brittain, S. M.;
Schultz, P. G. Nat. Chem. Biol. 2007, 3, 769.
(
(
(
(
(
(
(
20) Liu, C. C.; Schultz, P. G. Annu. Rev. Biochem. 2010, 79, 413.
21) Deiters, A.; Groff, D.; Ryu, Y.; Xie, J.; Schultz, P. G. Angew.
Chem., Int. Ed. 2006, 45, 2728.
22) Zhao, J.; Lin, S.; Huang, Y.; Zhao, J.; Chen, P. R. J. Am. Chem.
Soc. 2013, 135, 7410.
23) Gautier, A.; Deiters, A.; Chin, J. W. J. Am. Chem. Soc. 2011, 133,
124.
24) Groff, D.; Wang, F.; Jockusch, S.; Turro, N. J.; Schultz, P. G.
Angew. Chem., Int. Ed. 2010, 49, 7677.
25) Kang, J. Y.; Kawaguchi, D.; Coin, I.; Xiang, Z.; O’Leary, D. D.;
Slesinger, P. A.; Wang, L. Neuron 2013, 80, 358.
26) Hemphill, J.; Chou, C.; Chin, J. W.; Deiters, A. J. Am. Chem. Soc.
013, 135, 13433.
27) Gautier, A.; Nguyen, D. P.; Lusic, H.; An, W.; Deiters, A.; Chin,
J. W. J. Am. Chem. Soc. 2010, 132, 4086.
(
(
2
(
(
(
2
(
(
28) Baker, A. S.; Deiters, A. ACS Chem. Biol. 2014, 9, 1398.
(
29) Nguyen, D. P.; Mahesh, M.; Elsasser, S. J.; Hancock, S. M.;
Uttamapinant, C.; Chin, J. W. J. Am. Chem. Soc. 2014, 136, 2240.
30) Uprety, R.; Luo, J.; Liu, J.; Naro, Y.; Samanta, S.; Deiters, A.
ChemBioChem. 2014, 15, 1793.
31) Chen, S.; Chen, Z. J.; Ren, W.; Ai, H. W. J. Am. Chem. Soc. 2012,
34, 9589.
(
(
1
(
32) Bochet, C. G. J. Chem. Soc., Perkin Trans. 1 2002, 125.
(
33) Li, Y.; Sierra, A. M.; Ai, H. W.; Campbell, R. E. Photochem.
Photobiol. 2008, 84, 111.
34) Macdonald, P. J.; Chen, Y.; Mueller, J. D. Anal. Biochem. 2012,
21, 291.
35) Johannessen, C. M.; Boehm, J. S.; Kim, S. Y.; Thomas, S. R.;
(
4
(
Wardwell, L.; Johnson, L. A.; Emery, C. M.; Stransky, N.; Cogdill, A.
P.; Barretina, J.; Caponigro, G.; Hieronymus, H.; Murray, R. R.; Salehi-
Ashtiani, K.; Hill, D. E.; Vidal, M.; Zhao, J. J.; Yang, X.; Alkan, O.; Kim,
S.; Harris, J. L.; Wilson, C. J.; Myer, V. E.; Finan, P. M.; Root, D. E.;
Roberts, T. M.; Golub, T.; Flaherty, K. T.; Dummer, R.; Weber, B. L.;
Sellers, W. R.; Schlegel, R.; Wargo, J. A.; Hahn, W. C.; Garraway, L. A.
Nature 2010, 468, 968.
(
36) Wang, X.; Pineau, C.; Gu, S.; Guschinskaya, N.; Pickersgill, R.
W.; Shevchik, V. E. J. Biol. Chem. 2012, 287, 19082.
37) Seong, J.; Lu, S.; Ouyang, M.; Huang, H.; Zhang, J.; Frame, M.
C.; Wang, Y. Chem. Biol. 2009, 16, 48.
38) Hemphill, J.; Govan, J.; Uprety, R.; Tsang, M.; Deiters, A. J. Am.
Chem. Soc. 2014, 136, 7152.
39) Lockless, S. W.; Muir, T. W. Proc. Natl. Acad. Sci. U.S.A. 2009,
06, 10999.
(
(
(
1
D
DOI: 10.1021/ja508597d
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX